Facebook was able to make an artificial intelligence feature that possibly detects the drug combinations that are efficient against cancer. These combinations could prevent and cure cancers.
Facebook’s innovation can predict the way drugs interact with each other inside cells quicker than existing methods. Thanks to this, the social media giant can now help medical experts find new drug combinations faster, allowing them to treat patients’ illnesses.
However, it is still criticized by some medical researchers across the globe, saying that it may not translate into results that will be useful in humans. Facebook AI Researcher and the Helmholtz Center in Munich Germany.
They worked together to developed Facebook’s new AI, which is considered to be the first easy-to-use AI model that can estimate how efficient different drugs will work in a human body. This innovation can speed up experts’ ability to uncover new treatments for diseases, such as cancer.
Fabian Theis, one of Helmholtz’s experts involved in the study, said that the new “drug research often takes half a decade to develop a compound.” On the other hand, Theis also explained that the new Facebook AI works by measuring how individual cells change in response to treatment from a particular set of drugs and recording those responses.
How effective is the new AI of Facebook
Researchers involved in the study explained that the new AI of the popular social media app can theoretically tackle cancer tumors, which vary from person to person and react differently to the same treatment. Eytan Ruppin, an expert from the U.S. National Cancer Institute, explained that AI works through different variables. These include the type of drug, what it is used in combination with, the dosage level, the time it is taken, and the type of cell it targets.
The social media app’s new AI can use this info to identify the effect of drug combination it hasn’t yet seen. The research team behind it says humans can’t make these kinds of predictions. They added that if various medical experts were given a pool of 100 different drugs, and asked to choose five to be given in three different doses, they would have more than 19 billion drug combinations, which makes it harder for them to identify the efficient ones.
As of the moment, the new AI is still being tested by the giant social media app. They are conducting experiments to find if the innovation is really effective when it comes to predicting known combinations of drugs.
But, the involved experts confirmed that the current results revealed that the new AI of this social media app has an accuracy of 90%. Although this the case, they still need to conduct further tests to confirm since they’ll never know if its accuracy could increase more than 90%.
Once it is completed, the social media app company will release its new technology as an open-source tool for the research community to use and develop. David Lopez-Paz, a researcher at the company’s AI Research department, said that “a single model can be trained in a few hours on a single machine.”